STAI: scoring, cutoffs & interpretation

01 - What it measures

What this scale measures.

The State-Trait Anxiety Inventory (STAI) measures two distinct things that most anxiety scales blur together: state anxiety - how anxious the person feels right now, in this moment - and trait anxiety, the person's general disposition to respond to situations with anxiety. It comprises 40 self-report items in two 20-item subscales, each rated on a four-point intensity or frequency scale and each yielding a score from 20 to 80.\n\nThe state-trait distinction is the instrument's defining contribution. A patient facing surgery tomorrow may score high on state anxiety with an unremarkable trait score; a patient with generalised anxiety disorder will usually carry an elevated trait score whatever the day brings. Reading the two subscales together tells you whether you are looking at a situational reaction or a stable characteristic.

02 - Origin & purpose

Where it comes from.

The STAI was developed by Charles Spielberger and colleagues, first published as Form X in 1970 and revised as Form Y in 1983. The Form Y revision replaced items that overlapped with depression, improving the scale's ability to distinguish anxiety from low mood, and is the version in standard use today. It is among the most widely used and cited anxiety measures in clinical research, with adaptations in dozens of languages.\n\nUnlike the free instruments that dominate modern screening, the STAI predates the open-access era and remains commercially controlled. It is published and licensed by Mind Garden, Inc., which charges per administration. That commercial status, together with its 40-item length, has pushed routine clinical screening towards briefer free tools, while the STAI retains a strong position in research and in settings where the state-trait distinction matters.

03 - Scoring & cutoffs

How scoring works.

Each of the two subscales is scored separately. Items are rated 1 to 4; roughly half are anxiety-absent statements that are reverse-scored. Each subscale therefore ranges from 20 to 80, with higher scores indicating greater anxiety. There are no diagnostic cutoffs: the STAI was normed on working adults, students and military recruits, and interpretation is usually against those norms. A cut point of 39-40 on the state subscale is commonly cited for clinically significant state anxiety, with research in older adults suggesting a higher threshold of 54-55. Treat these as conventions from the literature rather than validated diagnostic thresholds.

04 - Validation evidence

How well it performs.

Internal consistency is high: Cronbach's alpha across studies and translations typically falls between 0.86 and 0.95. Test-retest reliability behaves exactly as the state-trait theory predicts: trait scores are stable (0.65-0.75 over two months, 0.73-0.86 over three to fifteen weeks in other samples), while state scores are deliberately unstable (median around 0.33), because state anxiety is supposed to change with circumstances. A low state retest coefficient is a design feature, not a flaw. The main psychometric criticism is discriminant validity: even Form Y correlates substantially with depression measures, so an elevated score cannot by itself distinguish anxiety from mixed anxiety-depression.

05 - How it compares

How it compares to the alternatives.

06 - When to use it

Right tool, wrong tool.

07 - Confidence & precision

Reading the score with care.

A standard error of measurement is not routinely reported for the STAI in a way that generalises across settings, because interpretation is norm-referenced and norms differ by age, sex and population. No consensus minimal clinically important difference has been established. In practice this means treating between-visit changes conservatively: interpret trait-score movement only when it is large and sustained, and expect state scores to move with context - that is what they are for.

08 - Limitations

What it cannot tell you.

The chief limitation is discriminant validity: STAI scores correlate substantially with depression measures, and even the Form Y revision does not fully separate the constructs, so elevated scores are ambiguous in mixed presentations. It is long by modern screening standards, and there are no validated diagnostic cutoffs - the commonly cited 39-40 convention derives from normative comparisons rather than diagnostic accuracy studies.\n\nThe norms themselves are decades old and drawn largely from US samples. Finally, the licensing cost and prohibition on reproduction make it impractical for routine use in most publicly funded services when free alternatives with validated cutoffs exist.

09 - Licensing, explained

How licensing works.

Copyright Charles D. Spielberger; published and licensed by Mind Garden, Inc. with a per-administration fee. Item text may not be reproduced. Free alternatives with validated cutoffs: GAD-7 and GAD-2.

Common questions from clinicians

FAQ

References

Source literature.

1. [1]Spielberger CD, Gorsuch RL, Lushene R, Vagg PR, Jacobs GA Manual for the State-Trait Anxiety Inventory (Form Y). Palo Alto, CA: Consulting Psychologists Press; 1983 ↩
2. [2]Julian LJ Measures of anxiety: State-Trait Anxiety Inventory (STAI), Beck Anxiety Inventory (BAI), and Hospital Anxiety and Depression Scale-Anxiety (HADS-A). Arthritis Care Res. 2011;63(S11):S467-S472 ↩
3. [3]Marteau TM, Bekker H The development of a six-item short-form of the state scale of the Spielberger State-Trait Anxiety Inventory (STAI). Br J Clin Psychol. 1992;31(3):301-306 ↩
4. [4]Mind Garden, Inc. State-Trait Anxiety Inventory for Adults - licensing ↩